Applications and explorations of CRISPR/Cas9 in CAR T-cell therapy.

Abstract Chimeric antigen receptor(CAR) T-cell therapy has shown remarkable effects and promising prospects in patients with refractory or relapsed malignancies, pending further progress in the next-generation CAR T cells with more optimized structure, enhanced efficacy and reduced toxicities. The clustered regulatory interspaced short palindromic repeat/CRISPR-associated protein 9 (CRISPR/Cas9) technology holds immense promise for advancing the field owing to its flexibility, simplicity, high efficiency and multiplexing in precise genome editing. Herein, we review the applications and explorations of CRISPR/Cas9 technology in constructing allogenic universal CAR T cells, disrupting inhibitory signaling to enhance potency and exploration of safer and more controllable novel CAR T cells.


Introduction
Cancer immunotherapy is the fourth mainstream treatment after surgery, chemotherapy and radiotherapy. Adoptive T-cell immunotherapy, particularly chimeric antigen receptor (CAR) T cell therapy, has revolutionized cancer therapy especially after the FDA approval of Kymriah and Yescarta (CD19-directed CAR T cells in B-cell leukemia and lymphoma) [1][2][3]. CARs are synthetic receptors typically containing an antibodyderived target-binding extracellular domain, a hinge region, a transmembrane domain and an intracellular signaling moiety capable of activating T cells [4,5]. T cells programmed with CARs can specifically recognize and kill antigen-expressing cells without the restriction of major histocompatibility complex (MHC). Clinical data has demonstrated that CAR T-cell therapy can induce durable complete remissions (CRs) in patients with a variety of hematologic and solid cancers, especially in relapsed/refractory acute lymphoblastic leukemia (ALL) and multiple myeloma with striking response rates of 80-100% [6][7][8]. Despite of promising efficacy of CAR T-cell therapy, there are several challenges awaiting for solutions, such as insufficient quantity and poor quality of autologous T cells, CAR T cell exhaustion and tumor suppressive microenvironments, potential self-killing and uncontrollable proliferation.
Optimization of the CAR T designs is supposed as one of the main tracks to tackle these limitations. The first generation of CAR T cells with only CD3 zeta intracellular chain was found to have modest proliferative and cytotoxic capacity [9][10][11][12]. The second generation of CARs contains a single costimulatory domain (CD28 or 4-1BB), proven to attain an improved efficacy and in vivo survival, whereas the third generation has two or more costimulatory domains (CD28, 4-1BB, ICOS or OX40), not superior to the second generation [13][14][15]. More functional elements are considered to be added to the next generation of CARs, like interleukins genes to increase potency, chemokine receptors genes to improve T-cell trafficking and on-off switches or suicide genes to enhance safety and controllability [16][17][18]. The structures and features of every generation of CAR-T are shown in Table 1.
The development of genomic editing technologies opens a window to accelerate the fourth generation of CAR T cells. There are currently three major genomic editing technologies, including zinc-finger nucleases (ZFNs), transcription activatorlike effector nucleases (TALENs) and clustered regulatory interspaced short palindromic repeat/CRISPR-associated protein 9 (CRISPR/Cas9) [19][20][21]. Although ZFNs and TALENs have been applied to engineer T cells in clinical trials, the recognition of targetable DNA sequences is based on complicated protein conformation, a pair of Zn-finger binding domains or a pair of TALE DNA binding domains, accompanying with complex designs and relatively low gene-editing efficiencies [22,23]. CRISPR/Cas9, directed by a small guide RNA (sgRNA) to the target site, has become the most popular and developed of these tools due to its simplicity, flexibility, high efficiency and multiplexable genome editing capabilities [24][25][26]. A sgRNA-guided Cas9 nuclease induces a DNA double-stranded break at targeted genomic locations, subsequently repaired by non-homologous end joining (NHEJ) or homology-directed repair (HDR). NHEJ, an error prone repair pathway, can result in insertions or deletions of small nucleotide sequences and HDR can knock-in relatively large gene segments in the presence of a homology repair template at the site of interest [27][28][29]. Therefore, the combination with CRISPR/Cas9 technology will further expand the landscape of T-cell engineering. Besides knock-in of functional genes, such as interleukins and suicide genes, to product next-generation CAR T cells, other strategies comprises knock-out of endogenous genes, such as TCRs and MHCs, to develop 'off-the-shelf' universal CAR T cells [30], disruption of inhibitory receptors (such as PD-1 and TGF beta receptor) to ameliorate suppressive microenvironments [31,32], integration of the CAR cassette into the specific gene locus(such as TRAC and TET2) to improve efficiency and safety [33,34], deletions of target genes to avoid selfkilling of CAR T cells [35]. CRISPR/Cas9 technology is unveiling a new era for CAR T-cell therapy. All gene-edited CAR T cells discussed here are shown in Table 2.

Production of allogeneic universal CAR T cells
Although currently widespread-used autologous CAR T cells have shown promising results in cancer therapy, limitations exist. Almost 10-15% of enrolled patients were unable to receive infusions of CAR T cells because of poor quality and insufficient quantities of autologous T cells unavailable for manufacturing or rapid disease progression and even death before successful production of certain amount of CAR T cells [1][2][3]. A UPenn team recently reported a patient relapsing after infusion of anti-CD19 CAR T cells with CD19-negative leukemia that aberrantly expressed the anti-CD19 CAR because the CAR gene was unintentionally introduced into a single leukemic B cell during T-cell manufacturing [36]. The development of universal 'off-the-shelf' CAR T cells from healthy donors can circumvent the constraints and potentially be the mainstream direction in the future. The major barriers of such universal CAR T cell products are graft-versus-host disease (GVHD) and rejection of the infused allogeneic T cells. Endogenous αβ T cell receptors (TCRs) on adoptively transferred donor lymphocytes can recognize alloantigens in human leukocyte antigen (HLA) mismatched recipients resulting in GVHD; conversely, recognition of foreign HLA molecules on donor T cells may lead to rejection. ZFNs and TALENs were successfully used to knock-out TCRα constant (TRAC) and TCRβ constant (TRBC) to generate TCRnegative CAR T cells to prevent GVHD without compromising CAR-mediated cytotoxicity [37,38]. Previous researches demonstrated that genetic knock-out of either TRAC or TRBC loci was sufficient to eliminate expression of αβTCR on the T cell surface [39]. The Cellectis firstly reported the generation of TALENedited allogeneic universal anti-CD19 CAR T(UCART19) cells in which TRAC and CD52 genes were knocked out [40]. CD52 disruption in the CAR T cells allowed effective targeted depletion of patients' autologous T cells using an anti-CD52 antibody (alemtuzumab). The first-in-man application of the products was two infants with high-risk CD19-positive ALL who achieved molecular remission after receiving the infusion of UCART19 cells and attained successful bridge-to-transplantation [41,42]. The remarkable results led to two clinical trials of UCART19 cells: CALM trial in adults and PALL trial in pediatric patients (NCT02746952 and NCT02808442). Pooled data of 20 patients showed acceptable and manageable safety with 15% (3/20) of severe cytokine release syndrome (CRS) and 10% (2/20) of G1 cutaneous acute GVHD as well as promising efficacy with 88% (14/16) of CR or CR with incomplete blood count recovery (CRi) and 86% (12/14) of minimal residual disease-negative [43]. A MSKCC group showed that directing a CD19-specific CAR to the TRAC locus using CRISPR/Cas9 technology not only minimized the risks of insertional oncogenesis and TCR-induced GVHD, but also enhanced T-cell potency and delayed T-cell exhaustion [33]. ZFNs were also used to target the HLA-A locus to permanently and completely eliminate HLA-I expression in primary and genetically modified human T cells used in clinical trials to evade rejection [44]. In addition, elimination of HLA heavy chains or beta-2-microglobulin (B2M), the non-polymorphic subunit of HLA-I complex, would prevent rapid rejection of allogeneic cells [45]. However, ideal universal CAR T cells should be silenced both TCR and HLA to avoid GVHD and rejection without reducing persistence and cytotoxicity in vivo. CRISPR/Cas9 has an obvious advantage in simultaneously multiplex and highly efficient genomic editing compared with ZFNs and TALENs. CRISPR/Cas9 was readily applicable to generate double-knockout (B2M and TRAC, DKO) UCART19 cells with as similar safety and efficacy as wild-type anti-CD19 CAR T cells in preclinical studies [30]. One-shot CRISPR protocol for multiplex genome editing by incorporating multiple gRNA cassettes into a single CAR lentiviral vector was developed to generate DKO UCART19 cells [46]. By combining the lentiviral delivery of CAR with CRISPR RNA electroporation to co-introduce RNA encoding the Cas9 and gRNAs targeting endogenous TCR and B2M concurrently, an improved editing efficiency(∼80%) was acquired to construct the DKO UCART19 cells that showed as potent antitumor activities as non-gene-edited CAR T cells both in vitro and in animal models [47]. However, the issue of whether such HLA-I negative CAR T cells will be the target of NK cells should be considered.
Administering an anti-NK cell depletion antibody or engineering T cells with HLA-E are potential solutions to circumvent NKmediated rejection [44,48].
Recent advances in gene-editing technology, especially CRISPR/Cas9, allow for the production of universal CAR T cells starting from healthy donor in which the manufacture and quality of T cells can be preselected as well as GVHD and rejection can be avoided. The CRISPR/Cas9-modified universal CAR T cells need to be further tested for the safety and efficacy in clinical studies and there are currently six relevant ongoing clinical trials ( Table 3).

Disruption of inhibitory signaling molecules
The function of T cells was proven to play a significantly important role in the therapeutic effect of CAR T cells [49]. However, T cells are exposed to persistent antigen in patients with malignant tumors, resulting in T-cell exhaustion [50]. Exhausted T cells lose robust effector functions and express multiple inhibitory receptors, such as programmed cell death 1 (PD-1), cytotoxic T-lymphocyte antigen 4 (CTLA-4), domaincontaining protein-3 (TIM-3) and lymphocyte-activated gene-3 (LAG-3), which inhibit T-cell proliferation and cytokine production leading to immune escape [51]. The inhibitory pathways also contribute to suppressive tumor microenvironment, a major barrier of CAR T-cell therapy in solid tumors. Immune checkpoint inhibitors, anti-PD-1/PD-L1 and anti-CTLA-4 antibodies, have shown promising clinical results and been approved by the FDA [52]. Thus, disruption of multiple inhibitory factors is expected to improve the potency of CAR T cells. Recent studies suggested that anti-CD19 CAR T cells with CRISPR-mediated triple-knockout of the TRAC/TRBC, B2M and PD-1 genes displayed stronger antitumor functions in contrast to DKO UCART19 cells in vitro and in animal models [30,47]. High-fidelity Cas9s with the oneshot platform showed the feasibility of generating PD-1 and CTLA-4 dual inhibitory pathway-resistant DKO UCART19 cells by simultaneous disruption of quadruple genes [46].
The Fas receptor is a member of the tumor necrosis factor α (TNF-α) family of death receptors that mediate cell death [53]. Researches demonstrated that CAR T cell activity was attenuated due to cell Fas-FasL-dependent activation-induced cell death (AICD) [54]. Ren et al. [46] also utilized CRISPR/Cas9 technology to generate Fas-resistant universal CAR T cells that observed elevation of AICD resistance and prolonged survival.
Transforming growth factor-β (TGF-β) represses effector Tcell activities through binding the TGF-β receptors (TGFBRI and TGFBRII) to induce heterodimerization of the respective receptors and phosphorylation of the major TGF-β signal mediators SMAD2 and SMAD3, resulting in reduced cytokine production, cytotoxicity and amplification [55]. TGF-β also drives T-cell differentiation into regulatory T cells (Tregs) [56]. Thus, inhibiting TGF-β signaling, a potent immunosuppressive factor in a variety of solid tumors, has the potential to improve the immunosuppressive milieu. Previous studies demonstrated that TGF-β pathway could be blocked by using a dominant-negative TGFBRII (dnTGF-βRII), which lacked the intracellular domain necessary for downstream signaling [57]. Foster et al. [31] used a clinical grade retrovirus vector to construct dnTGF-βRII-expressing human antigen-specific cytotoxic T lymphocytes (CTLs) and found that TGF-β-resistant CTLs had a functional advantage over unmodified CTLs in the TGF-β-secreting lymphoma [31]. The clinical trial (NCT00368082) showed that TGF-β-resistant CTLs could safely expand and persist in patients with Hodgkin lymphoma without lymphodepleting chemotherapy and induced complete responses [58]. It was testified that adding dnTGF-βRII to PSMA-targeted human CAR T cells promoted T-cell proliferation and augmented prostate cancer eradication [59]. Chang et al. recently described a novel TGF-β CAR containing a scFv based on the sequences of TGF-β-neutralizing antibodies, demonstrating the ability to not only inhibit endogenous TGF-β signaling but also convert TGF-β into a stimulant of T-cell growth [60]. Above results support the potential value of the countermeasure of using CRISPR/Cas9 technology to generate TGF-β-resistant CAR T or UCART cells to improve potency of engineering T cells in solid tumors.

Exploration of safer and more controllable novel CAR T cells
Albeit unprecedented efficacy of CAR T-cell therapy, it is accompanied by serious and even life-threatening toxicities, including CRS, on-target/off-tumor toxicity, neurotoxicity, macrophage activation syndrome/ hemophagocytic lymphohistiocytosis and tumor lysis syndrome, which need to be paid more attention [61]. The most significant and common toxicity of CAR T-cell therapy is CRS, an inflammatory syndrome caused by multiple cytokines, including interferon γ , interleukin (IL)-1, IL-2, IL-4, IL-6, IL-8, IL-10, TNF-α and granulocyte/macrophage colony-stimulating factor (GM-CSF), produced by the CAR T cells themselves and by other cells [62]. Tocilizumab, IL-6 receptor blockade, was approved by FDA for treatment of CAR T cell-induced severe or life-threatening CRS [63]. Thus, blocking relevant cytokines signaling is a hopeful strategy to ameliorate the dilemma and CRISPR/Cas9 can effectively knock-out related molecules. Sterner et al. [64] described CRISPR/Cas9 mediated knock-out of GM-CSF and showed that GM-CSF-negative CAR T cells produced less GM-CSF without weakening antitumor activity in vivo compared to wildtype CAR T cells. Single or combined knock-out of other critical relevant cytokines in CAR T cells using CRISPR/Cas9 are needed to be further explored. Long-lasting B cell aplasia is a classical on-target/off-tumor toxicity of anti-CD19 and anti-CD20 CAR T-cell therapy [65,66]. The insert of safety switches gene into CAR vector is a feasible method to terminate the effects without jeopardizing clinical responses. Diaconu et al. [67] demonstrated that the iC9 safety switch eliminated CD19-specific CAR T cells in a dose-dependent manner in a humanized mouse model, allowing either a selective containment of CAR T expansion in case of CRS or complete deletion on demand granting normal B-cell reconstitution.
There are two reported cases indicating the risks of unexpected situations in the manufacture of CAR T cells and potential carcinogenicity in vivo. One CD19-negative relapsed patient after CD19-targeted CAR T cell therapy was found that the CAR gene was unintentionally introduced into a single dominantly-proliferative leukemic B cell and its product bound in cis to the CD19 epitope on the surface of leukemic cells, masking it from recognition by CAR T cells [36]. Another case was a 78-year-old man with advanced relapsed/refractory chronic lymphocytic leukemia who obtained CR after the second infusion. Unexpectedly, 94% of CAR T cells at the peak of the response originated from a single clone in which lentiviral vector-mediated insertion of the CAR transgene disrupted the methylcytosine dioxygenase TET2 gene [34]. Therefore, there is a need to incorporate inducible safe switches or suicide genes into the CAR T cells, which can provide a means to eliminate the CAR T cells in case of unexpected toxicities. Hoyos et al. [68] generated a novel anti-CD19 CAR construct that incorporates the IL-15 gene and an inducible caspase-9(iC9)-based suicide gene and >95% of transgenic cells could be efficiently eliminated within 24 h upon pharmacologic activation of the suicide gene [68]. Adding inducible safe switches or suicide genes to generate more controllable and safer CAR T cells will be widespread-used by multiplexed CRISPR/Cas9 technology. There are three ongoing clinical trials (NCT02107963, NCT01822652 and NCT02439788) incorporating the iC9 construct into CAR T-cell products to provide a method to eliminate autologous CAR T cells in case of potential off-target toxicity.

Other applications
CAR T-cells therapy has an obvious barrier in acute myeloid leukemia (AML) because myeloid-directed immunotherapy will eradicate normal as well as malignant cells, leading to bone marrow failure, as has been shown in several preclinical studies of CD33 or CD123 directed CAR T cell therapy [69,70]. Several groups developed a novel approach to circumvent the problem with potent anti-CD33 CAR T cells followed by infusions of CRISPR/Cas9-modified CD33-konckout normal hematopoietic stem cells (HSCs), thus allowing persistent antigen-specific cytotoxicity along with reconstitution of effective hematopoiesis [35,71,72]. Extending the success of CAR T cells to T-cell malignancies is also problematic because most target antigens are expressed on both normal and malignant cells, resulting in CAR T-cell fratricide. CD7 is a transmembrane protein highly expressed in T-cell acute leukemia (T-ALL) and largely confined to T cells and natural killer cells. Studies showed that CD7specific CAR T cell impaired expansion due to self-killing of the CAR T cells. Diogo et al. [73] explored that targeted disruption of the CD7 gene using CRISPR/Cas9 prior to CAR expression minimized fratricide in T cells and allowed the expansion of the CD7-knock-out anti-CD7 CAR T cells with robust antitumor activity for preclinical and potential clinical application. Hence, the CRISPR/Cas9 system can be applied to disrupt the targeted antigens to avoid self-killing of the CAR T cells and broaden the therapeutic index.

Conclusion and outlooks
The unprecedented responses of CAR T cells in advanced malignancies promote the rapid growth of the therapeutic approach and the development of the smarter and commercialized CAR T cells is an inevitable mainstream trend, such as a split, universal and programmable CAR system to prevent relapse, mitigate overactivation and enhance specificity [74]. CRISPR/Cas9 genomic editing technology holds promising explorations and applications to create the next-generation CAR T-cell products, including universal CAR T cells by disrupting endogenous TCR and HLA, more potent CAR T cells by ablating inhibitory modulators, more controllable CAR T cells by adding inducible safe switches or suicide genes and novel CAR T cells by knock-out of the targeted antigens to avoid self-killing.
However, the gene-editing specificity and efficiency of CRISPR/Cas9 technology are of significant importance in therapeutic application. The first concern of CRISPR/Cas9 gene editing is off-target effects, which introduce random mutations, hence activating oncogenes or impacting tumor-suppressor genes to unintentional deleterious consequences [75]. Multiple strategies, such as careful selection of the target site, optimized sgRNA design and Cas9 activity, prior off-target detection assays, have been attempted to minimize the safe risks of off-target effects [76][77][78]. Attempts to increase HR frequencies using HR enhancers or NHEJ inhibitors are currently ongoing and may further promote precise gene engineering [79,80]. Another challenge for therapeutic gene editing is efficient and nontoxic delivery into CAR T cells. There are three main methods to deliver CRISPR/Cas9 system, including a DNA plasmid-based system, an all-RNA-based system and a Cas9 ribonucleoprotein complex as delivery [81]. Viral vectors with high efficiency and potential hazards, such as mutagenesis, immunogenicity and off-target effects, are widely applied for donor DNA delivery and electroporation has emerged as new method to deliver CRISPR/Cas9 elements with safety, simplicity and flexibility [82]. Viral and non-viral vectors have specific merits and beneficial combinations of different delivery means are being explored to ensure efficiency and safety [33]. As technical progresses to reduce off-target effects and improve delivery efficiency, CRISPR/Cas9 technology provides an extraordinary potential to construct novel CAR T cells and streamlines the burgeoning realm of immunotherapy.

Key Points
• CAR T-cell therapy has shown promising responses in both hematologic and solid cancers. However, there are some limitations awaiting for solutions, such as insufficient quantity and poor quality of autologous T cells, CAR T cell exhaustion and tumor suppressive microenvironments, potential self-killing and uncontrollable proliferation.
• Genomic editing technologies, especially CRISPR/Cas9 with flexibility, simplicity, high efficiency and multiplexing open a window to develop next-generation CAR T cells.
• CRISPR/Cas9 genomic editing technology holds promising explorations and applications to create nextgeneration CAR T cell products, including universal CAR T cells by disrupting endogenous TCR or HLA, more potent CAR T cells by ablating of inhibitory modulators and more controllable CAR T cells by adding inducible safe switches or suicide genes.
• CRISPR/Cas9 technology is unveiling a new era for CAR T cell therapy.

Conf lict of interest
The authors declare that they have no conflict of interests.

Funding
The work was supported by grants from the Key Special Project of "Research on Prevention and Control of Major Chronic Noninfectious Diseases" (No.2019YFC1316203) and the National Natural Science Foundation of China?(No. 81873434).